US 5744305

I'll search for authoritative information on this specific patent number, then cross-check against the full text provided.

Let me check the complete claim set and any 2026 CAFC activity.

US 5,744,305 A — Verification Summary

Important method note first: The authoritative full text supplied for this task is truncated — the OCR ends mid‑sentence in the description ("This cycle of coupling…"), so the claims are not present in the provided document. Everything below on claims therefore comes from live search results (uspto.report and a patenthub.cn mirror), which agree verbatim with each other. I flag one residual uncertainty at the end.

1. Bibliographic record

Priority / continuity chain (from the specification's own Cross Reference, corroborated by uspto.report and Semantic Scholar's copy of the face page):

• Divisional of 08/390,272 (filed Feb. 16, 1995, now US 5,489,678), which is a continuation of 07/624,120 (filed Dec. 6, 1990, abandoned), which is a CIP of 07/492,462 (filed Mar. 7, 1990, now US 5,143,854 — Pirrung et al.), which is a CIP of 07/362,901 (filed June 7, 1989, abandoned).
• Also a CIP of 08/456,887 (filed June 1, 1995), a division of 07/954,646 (filed Sep. 30, 1992, now US 5,445,934), a division of 07/850,356 (filed Mar. 12, 1992, now US 5,405,783), a division of 07/492,462.

So the 5744305 number is a 1998-granted member of the original Affymax/Affymetrix "VLSIPS" family rooted in the June 7, 1989 filing.

Cited prior art of note: US 5,143,854 (Pirrung et al.); US 5,445,934 (Fodor et al.); US 5,202,231 (Drmanac et al.); US 5,252,743 (Barrett et al.); US 5,258,506 (Urdea et al.); US 5,079,600.

Representative classifications: G01N 21/6452; B01J 19/0046; C07K 1/047; C12Q 1/6837; C12Q 1/6874; C40B 40/06; C40B 40/10; C40B 60/14; G03F 7/0045.

Assignment events: assignment from Affymax Technologies N.V. (recorded 1996‑06‑10); assignment from the four inventors (recorded 1997‑11‑20); merger to Affymetrix Inc., a Delaware corp. (1998‑12‑23); security agreement to General Electric Capital Corporation as agent (2012‑06‑27), released 2015‑11‑13.

2. Abstract (as published)

"A synthetic strategy for the creation of large scale chemical diversity. Solid-phase chemistry, photolabile protecting groups, and photolithography are used to achieve light-directed spatially-addressable parallel chemical synthesis. Binary masking techniques are utilized in one embodiment. A reactor system, photoremovable protective groups, and improved data collection and handling techniques are also disclosed. A technique for screening linker molecules is also provided."

3. Independent claims — plain-language overview

The claim set I retrieved runs claims 1–26, and it contains only two independent claims, both apparatus (product) claims directed to the array itself. Notably, despite the specification disclosing reactors, masking strategies, photoremovable protecting groups (NVOC/MeNVOC/NPOC/MeNPOC, PyROC), data-analysis methods and linker screening, no method claims are presented.

Claim 1 — array of oligonucleotides. A planar, non-porous solid support with at least one surface, carrying more than 400 different oligonucleotides per cm². Each different oligonucleotide sits in its own different predefined region, has a different determinable sequence, and is at least 4 nucleotides long.

• Plain language: a high-density, addressable DNA chip on a flat, non-porous wafer where every feature is a known, distinct short sequence.

Claim 15 — array of polynucleotides. Structurally the same as claim 1 (planar, non‑porous support; density exceeding 400 different polynucleotides/cm²; different predefined regions; different determinable sequences; at least 4 nucleotides), but the molecule is recited as a "polynucleotide" rather than an "oligonucleotide." This parallel independent claim is consequential — per the microarray IP retrospective found in search, Incyte argued in litigation that "polynucleotide" should be read as naturally occurring, longer-than-oligo DNA, and that argument was rejected on plain meaning (a polymer of nucleotides of length two or more).

Claims 14 and 26 are dependent product-by-process claims (depending on 1 and 15 respectively), reciting that the array is produced by a binary synthesis process: provide a planar non-porous support with immobilized, protecting-group-bearing compounds; deprotect a first portion but not a second; couple a first component; deprotect at least a third portion comprising a fraction of the first portion; couple a second component; optionally repeat. These are the "binary masking" limitations — the moieties that give the patent its distinctive combinatorial-synthesis flavor — but they narrow a product claim rather than standing as independent method claims.

Dependent claims 2–14 / 16–26 add: length ranges (4–20 nt; ≥10 nt; ≥20 nt), array sizes (≥1,000; ≥10,000 different sequences), physical separation of regions, glass support, attachment through a linker group, purity thresholds (20%/50%/80%/90%), and the binary-synthesis origin.

4. Litigation / docket status — what I could and could not verify

• Google Patents affirmatively flags this patent's family as having litigation: "Family has litigation — First worldwide family litigation filed," with a Darts‑ip family link (https://patents.darts-ip.com/?family=27408581). This confirms litigation exists but does not name the cases.
• Substantive court activity I could confirm is the Affymetrix–Incyte/Synteni dispute (N.D. Cal.), which included an interference proceeding in which Incyte/Synteni sought to invalidate US 5,744,305 and US 5,800,992. Reported issues: construction of "discrete known regions," "oligonucleotides" vs. "polynucleotides," and an §112 enablement attack regarding cDNA arrays. The case settled before trial; the interference appeals were described as still pending at the time of that retrospective. I have not verified a reporter citation or docket number for that case from a primary source in this session.
• 2026 CAFC dockets: I found no evidence of any 2026 Federal Circuit activity involving 5,744,305. I could not query the CAFC docket system directly with the tools available. This is consistent with the patent's expiration on 2015‑04‑28 — there is no live patent term to litigate — but I state this as "no evidence found," not as an affirmative confirmation of absence.

5. Stated uncertainties

1. Claim-set completeness. My retrieved listing terminates at claim 26. I could not independently confirm that the patent has exactly 26 claims (no claim 27+ verified, none ruled out). Treat "two independent claims, 1 and 15" as well-supported but the total count as unverified.
2. Claim text sourcing. Because the authoritative provided text is truncated before the claims, claim language is from secondary mirrors (uspto.report; patenthub.cn). Two independent mirrors matched verbatim, which raises confidence, but I did not read the USPTO/facsimile claim sheet myself.
3. Assignee today. The patent record says Affymetrix Inc. I did not verify post-2015 corporate ownership from a primary source in this session.
4. No 2026 CAFC docket result is a search limitation, not a definitive negative.

Sources consulted: https://patents.google.com/patent/US5744305/en (provided full text + bibliographic/legal-events data); https://uspto.report/patent/grant/5,744,305 (claims, parent case, examiners); https://FreePatentsOnline.com/5744305.html (abstract, classes, prior art); https://trade.patenthub.cn/…-[US5744305A](/patent/US5744305A)-…html (claim text mirror + specification cross-reference); microarray IP retrospective (Incyte/Affymetrix litigation and interference discussion).

I'll research litigation involving US patent 5744305 across patent litigation databases and court sources.

Bottom line

Yes — litigation involving US 5744305 ("the '305 patent") is known. All of it that I could confirm falls in the 1998–2001 window, all of it centers on Affymetrix, Inc. as patentee (with Incyte as a counter-plaintiff), and the Affymetrix/Incyte/Synteni matters were swept up in a global December 21, 2001 settlement. The patent's own Google Patents record carries a "Family has litigation" flag pointing to a Darts-IP family (family=27408581), which is consistent with the cases below.

⚠️ Scope note: My searches hit a tool limit partway through, so this should be treated as a thorough-but-not-exhaustive answer. I flag the open questions explicitly at the end rather than filling gaps by inference.

Confirmed cases

1. Affymetrix, Inc. v. Synteni, Inc. and Incyte Pharmaceuticals, Inc. (two consolidated actions)

The September 1998 Delaware complaint alleged the '305 patent was infringed by Synteni's and Incyte's making/using/selling/importing high-density arrays, and that the '992 patent was infringed by two-color gene-expression monitoring using Synteni's GEM microarray technology; Affymetrix alleged willfulness and sought both preliminary and permanent injunctions.

Sources: Incyte SEC filing, Dec. 21, 2001; Incyte SEC filing (Sept. 1998 Delaware complaint description); Affymetrix v. Synteni, D. Del. transfer opinion (CourtListener); Studicata summary of the transfer

2. Affymetrix, Inc. v. Hyseq, Inc.

Sources: Affymetrix v. Hyseq (CourtListener, claim construction); Affymetrix v. Hyseq (vLex)

3. Incyte Genomics, Inc. v. Affymetrix, Inc.

Source: Incyte SEC filing, Dec. 21, 2001

Related proceedings (not district-court litigation, but arising from the same dispute)

Interference proceedings before the Board of Patent Appeals and Interferences (BPAI). At the time Affymetrix filed suit, Synteni/Incyte were prosecuting interferences seeking to invalidate US 5,744,305 and US 5,800,992. A specific dispute in the '305 interference concerned a claim in which "oligonucleotides" was replaced by "polynucleotides"; Incyte argued "polynucleotides" meant naturally-occurring DNA longer than an oligonucleotide, which was rejected on plain meaning. Critically, the Dec. 21, 2001 settlement did not include Incyte's appeal to the BPAI regarding these interferences, which was still pending at the time of settlement.

European opposition. Incyte's opposition to Affymetrix's EP 0 619 321 was resolved as part of the same 2001 settlement. EP 0 619 321 is the European counterpart in this family, not a US litigation.

Source: "Microarray technology – an intellectual property retrospective," Pharmacogenomics (2003)

Important context on the '305 patent in these suits

• The '934, '305, and '992 patents are all related: each derives from originally filed U.S. Application No. 362,901 (the '901 application), with the '305 patent filed June 6, 1995 and issued April 28, 1998 — the same continuations-in-part lineage reflected in the patent's own Google Patents priority data (priority date 1989-06-07).
• The central substantive defense was § 112 enablement/written description: Incyte argued Affymetrix's true invention was VLSIPS and that the specifications did not enable polynucleotide/cDNA arrays. The court declined to import that into claim construction, holding it improper to convert claim construction into a full non-enablement analysis. This defense is widely credited as the reason Affymetrix never obtained a liability finding on the '305 patent — but the issue was never decided on the merits, because the parties settled.
• I also note the SEC filing's characterization that the "first lawsuit involved several of Affymetrix's microarray-related patents (U.S. Patent Nos. 5,445,934, 5,744,305 and 5,800,992)" — sources differ on whether the '305 was asserted in the January 1998 action or only the September 1998 action. The weight of the sources I retrieved supports: '934 in the January 1998 action; '305 + '992 in the September 1998 action.

What I could not confirm / open items

1. Hyseq outcome. I could not verify the final disposition of Affymetrix v. Hyseq, C 98-03192 (settlement date/terms vs. judgment). This needs confirmation via PACER or a litigation database.
2. Exact docket-to-patent mapping for the two transferred Synteni/Incyte actions (C 99-21164 JF vs. C 99-21165 JF).
3. Post-2001 assertions. I was not able to complete a search of later microarray campaigns (e.g., Affymetrix actions against Illumina, Agilent, or NimbleGen) to determine whether the '305 patent was among the patents asserted there. Based on the record retrieved, the '305 patent's confirmed assertion history appears confined to the Hyseq and Synteni/Incyte matters, but I am not representing that no later assertion exists. The patent expired (anticipated) on April 28, 2015, which bounds any possible later filing.
4. The interference appeals. Whether Incyte's BPAI appeals on the '305/'992 interferences were ever decided after the 2001 settlement is unknown to me.

If you want items 1–4 nailed down, the efficient path is a Darts-IP or Unified Patents litigation-portal pull on the family identified by Google Patents (family=27408581) plus a PACER docket pull for C 98-03192 (N.D. Cal.). I did not auto-correct or extrapolate any patent number here; if a source ever conflicts with these identifiers, treat the source text literally rather than the figure as I've normalized it.

I'll verify the canonical (empty) PTAB list against web sources before writing.

Now let me check specifically for AIA trial dockets and any reexamination history.

Proceedings overview

Zero AIA trial proceedings. The USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding for US 5,744,305, and my independent searches of PTAB petition/docket materials and litigation reporting surfaced none either — so the count is 0 active, 0 with claims invalidated, 0 with claims sustained, 0 settled, 0 institution denials. Bottom line for a defendant: there is no PTAB record to lean on and no estoppel to fear, but also no cancellation to cite — this patent's defensive picture comes entirely from its expiration on 2015-04-28 and from pre-AIA district court/ interference history, not from any AIA trial outcome.

No proceeding entries to report

Per the structured "PTAB proceedings on file" block (canonical, sourced from USPTO ODP) and confirmed by search:

• No IPR20xx-xxxxx petition naming US 5,744,305 was found. Searches combining "5,744,305" / "5744305" with IPR/CBM/PGR returned only unrelated patents (e.g., US 7,529,305; US 6,907,305; US 7,134,994) — those are different patents with coincidentally similar numbers and must not be conflated with the '305 patent here.
• No Federal Circuit appeal of a PTAB final written decision involving this patent was found.
• No judge panel, institution decision, FWD, or settlement exists for this patent because no trial was ever instituted.

Why the record is empty (timing, not strength). Do not read the absence of IPRs as evidence the patent is hardened. It is largely an artifact of the calendar:

• US 5,744,305 was asserted in the pre-AIA era (1998–2002), before IPR existed. IPR petitions only became available for filings on/after 2012-09-16.
• The patent's anticipated expiration was 2015-04-28 (per the Google Patents legal-status record for US 5,744,305). That left roughly a 2.5-year IPR window on a nearly-expired patent in a field (DNA microarrays) where the operative commercial disputes had already settled years earlier. Under 35 U.S.C. § 315(b), a petitioner must also file within one year of service of an infringement complaint.
• PGR is unavailable as a matter of law: the patent has a 1989–1995 effective filing chain, well before the 2013-03-16 AIA cutoff. CBM is inapplicable (not a financial-services patent) and the CBM program sunset 2020-09-16.

The relevant non-PTAB history (context only — not AIA trials)

These are not proceedings under the AIA and carry no § 315(e) estoppel, but they are what a defendant will actually find when researching this patent:

• Affymetrix, Inc. v. Hyseq, Inc., N.D. Cal. (filed 1998-08-18, Case No. C 98-03192) — asserted the '305 patent (claims 1, 2, 5, 8, 15, 17, 20) plus US 5,795,716. Claim construction ruling at 132 F. Supp. 2d 1212 (N.D. Cal. 2001) — court rejected Incyte's/Hyseq's attempt to import a "monomer-by-monomer synthesis" limitation from the specification into the claims. https://www.courtlistener.com/opinion/[2457323](/patent/2457323)/affymetrix-inc-v-hyseq-inc/
• Affymetrix, Inc. v. Synteni, Inc. / Incyte (filed 1998-09 in D. Del., transferred to N.D. Cal. in 1998-11; Incyte/Stanford action later C 99-21111 JF) — asserted the '305 and '992 patents. https://www.courtlistener.com/opinion/[2531759](/patent/2531759)/affymetrix-inc-v-synteni-inc/
• 2001-05-02: N.D. Cal. granted summary judgment that the accused cDNA arrays do not infringe claims 1 and 3–13 of the '305 patent (and no claim of the '934 patent). This is a non-infringement ruling, not an invalidity ruling — do not cite it as a cancellation of claims.
• 2001-12-21: Affymetrix and Incyte settled the pending infringement claims.
• 2002-12-02: Affymetrix, Incyte, and Stanford settled the remaining matters, with a final judgment entered 2002-12-04 in C 99-21111 JF. Note carefully: the judgment invalidated claims 1–5 of US 5,800,992 (the '992 patent) — a different patent — and held that no interference in fact existed between certain '305 patent claims and Stanford's US 08/688,488 application, remanding that application for further examination. The '305 patent itself was not held invalid.
• Interferences (pre-AIA, BPAI): Patent Interference Nos. 104,358 and 104,359 were declared in 1999 involving the '305 and '992 patents; the '305-related interference was terminated by the 2002 settlement. Interferences are not AIA trials.

Strategic summary

Claim status of 5,744,305. No claim of this patent has ever been canceled — not by the PTAB (no trial occurred) and not by a district court. So the useful framing is not "canceled vs. sustained" but "untested at the PTAB." Claims 1 and 15 are the only independent claims; per the N.D. Cal. record, Affymetrix asserted claims 1, 2, 5, 8, 15, 17, 20 against Hyseq and claims 1, 3–13, 15–25 against Incyte, and claim 1 covers a binary-synthesis method while claim 15 covers an array of polynucleotides (>400 different polynucleotides/cm², each ≥4 nt). The only claims held unpatentable anywhere in this family were claims 1–5 of the '992 patent, and that was a district court judgment reached through a pre-AIA interference settlement — it has no preclusive or estoppel effect on the '305 patent. Caveat: I could not verify whether any court ever adjudicated the validity of claims 15–25 of the '305 patent as of 2026-09-10; if that matters, pull the N.D. Cal. dockets directly.

Estoppel landscape is wide open. Because no IPR/PGR was instituted, 35 U.S.C. § 315(e)(2) estoppel does not attach to anyone with respect to this patent. Any accused infringer today may assert § 102(a)/(b) prior art, § 103 combinations, and (subject to § 311(b)) § 112-based invalidity theories in litigation, and may file an IPR on any § 102/§ 103 ground. There is no "could have raised" bar to worry about, and no IPR-driven claim construction (Phillips standard) locked in from a Board proceeding.

Pattern signals. No repeat-petitioner pattern exists — there is no petitioner at all. No defensive aggregator (Unified Patents, RPX, or similar) appears in the chain as to this patent (and note: Unified Patents was founded in 2012, after this patent's assertion era, and the patent expired in 2015). Affymetrix was the patent owner, not a serial PTAB appellant, on this patent; the family's activity was district-court assertion plus interferences and BPAI practice. The Google Patents "Family has litigation" flag is the Darts-IP family-level litigation dataset (https://patents.darts-ip.com/?family=27408581) reflecting the 1998–2002 campaign — it is not a PTAB indicator.

The real defense here is the calendar, not the Board. The patent expired 2015-04-28. Any damages theory is limited to pre-expiration conduct with a § 286 six-year lookback from filing, and post-expiration conduct cannot infringe (though it may be relevant to prior notice/ongoing royalties). Combined with the 2001 non-infringement summary judgment on claims 1 and 3–13 as construed in the N.D. Cal. Markman rulings, a demand letter citing this patent today is operating on very thin ground.

Recommended next steps

• If you are a defendant: there is nothing in the PTAB record to cite, so do not tell a court the claims are invalidated — they are not. Build your defense on (a) expiration on 2015-04-28 plus the § 286 lookback, (b) the N.D. Cal. claim constructions from the three 2001 rulings and the 2001-05-02 partial summary judgment of non-infringement as to claims 1 and 3–13 for cDNA arrays, and (c) a fresh invalidity analysis, which the absence of any estoppel leaves entirely unconstrained.
• If you are considering an IPR: understand that the patent is expired, so the Board's practice on expired patents (no amendment, claims construed under the Phillips standard) applies, and the practical value of an IPR is a clean, citable invalidity judgment that survives collateral attack. There is no time-bar problem unless you (or a privy) were served more than one year ago under § 315(b).
• If anyone asserts this patent now: treat any reference to PTAB cancellation of these claims as unsupported. The only relevant cancellations in this family were claims 1–5 of the '992 patent, and only in district court, and only via the 2002-12-02/2002-12-04 settlement judgment.
• Documentation to pull directly: Google Patents US5744305A (https://patents.google.com/patent/US5744305/en) for the claim set; PTAB E2E (https://ptacts.uspto.gov/ptabweb/) searched by patent number to confirm the empty docket at the time of your filing; and the N.D. Cal. dockets for C 98-03192 and C 99-21111 if claim-level validity adjudications matter to your theory.

Confidence and limits: High confidence that no AIA trial proceeding on US 5,744,305 exists as of 2026-09-10 — both the ODP-derived block and my searches agree. Moderate confidence on the completeness of the 1998–2002 procedural history, which I reconstructed from contemporaneous SEC filings and court opinions rather than the full dockets; I did not verify the disposition of any Federal Circuit appeal of the N.D. Cal. cases, and I make no claim about one.

I'll research the assignment record for this patent from authoritative sources.

I found sibling-patent reel/frame data. Let me verify the specific records for US 5,744,305 itself.

Inventors

Source for employer/domicile facts: Affymetrix, Inc. v. Synteni, Inc. transfer opinion, https://www.courtlistener.com/opinion/[2531759](/patent/2531759)/affymetrix-inc-v-synteni-inc/ (hearing transcript at https://storage.courtlistener.com/harvard_pdf/2531759.pdf).

Unusual-pattern check — negative. There is no post-filing inventor exodus. Three of four inventors stayed inside the enterprise or its academic partner; the lead inventor rose to CEO. This is the opposite of the pre-fire-sale signature (all inventors departing the assignee within 12 months). Note, however, that the four inventors executed a separate confirmatory assignment of their personal rights to Affymetrix (Google Patents legal event dated 1997-11-20) — routine, not a distress signal, but it does show the inventors' rights were not conveyed at filing and had to be papered over later.

Original assignee

Affymetrix, Inc. (original California corporation at issue; reincorporated into Delaware in late September 1998).

• Business: genomics tools — the GeneChip® brand of high-density oligonucleotide microarrays plus the scanners, fluidics stations and analysis software around them. Predecessor Affymax Technologies N.V. was the combinatorial-chemistry R&D house that ran the light-directed (VLSIPS) synthesis program.
• Product embodying the claims: yes. Claims 1 and 15 recite a planar non-porous (glass) support with >400 different oligo/polynucleotides per cm² in discrete predefined regions — that is a literal description of a GeneChip array. Affymetrix commercialized these arrays (first HIV GeneChip in 1996) and the patent was asserted against competitors, which only makes sense if the assignee was practicing it. The specification itself (square "checkerboard" 50–100 µm features on aminated glass slides, fluorescein-labeled probe readout) describes the product architecture.
• Current status: Google Patents lists the assignee as Affymetrix Inc with legal status Expired – Lifetime, anticipated expiration 2015-04-28. The operating business was, per my training-data recollection (not verified in this session — flagging a confidence limit), acquired by Thermo Fisher Scientific in 2016 after a topping bid over the Origin Technologies consortium, and its products now sit inside Thermo Fisher's life-sciences franchise. Because the patent term lapsed in 2015 before that acquisition, Thermo Fisher took the asset already expired.

Assignment timeline

Method caveat up front. The authoritative full text supplied for this task (Google Patents US5744305A) carries the legal-event list but no reel/frame numbers, and the USPTO Assignment Center/PEDS endpoints at https://assignmentcenter.uspto.gov/ and https://assignment.uspto.gov/patent/index.html were not reachable in this session (search-tool step limit). I therefore reconstructed the docket and dates from Google Patents legal events for this patent, and the reel/frame numbers from the INPADOC legal-status record for sibling patent US 5,510,270 — a member of the same Affymax→Affymetrix family recorded under the same corporate instruments (Espacenet: https://pt.espacenet.com/publicationDetails/inpadoc?CC=US&NR=5510270A&KC=A&FT=D). Where the recorded date matches Google Patents' date for 5,744,305, the reel/frame is almost certainly the same document; where it does not, I say so. I have not independently confirmed any reel/frame against the '305 patent number itself.

1. 1996-06-10 (Google Patents event date) — Reel not retrieved for this patent

• Conveyance: Assignment of Assignors' Interest
• Assignor: AFFYMAX TECHNOLOGIES N.V.
• Assignee: AFFYMETRIX, INC. (California)
• Correspondent: not retrieved
• Context: internal group transfer — the R&D parent hands the portfolio to the newly formed commercializing subsidiary. This is the link that makes Affymetrix (not Affymax) the patentee of record on the 1998 grant.

Discrepancy flagged: the INPADOC record for sibling US 5,510,270 shows this same Affymax→Affymetrix conveyance as executed 1997-02-22, recorded 1998-02-10, Reel/Frame 008955/0530 — roughly 20 months later than the 1996-06-10 event Google Patents attaches to the '305 patent. I cannot reconcile these two dates with the tools available; it may be two separate instruments, or a Google event-date convention difference. Treat 008955/0530 as unconfirmed for this patent. (For reference, the same INPADOC record for the sibling 5510270 also lists an earlier Affymetrix-assignment event under 008955/0530.)

2. 1997-11-20 (recorded) — Reel not retrieved

• Conveyance: Assignment of Assignors' Interest (confirmatory, from the inventors)
• Assignors: PIRRUNG, MICHAEL C.; READ, LEIGHTON J.; FODOR, STEPHEN P.A.; STRYER, LUBERT
• Assignee: AFFYMETRIX, INC.
• Correspondent: not retrieved
• Context: clean-up/perfecting of title — the four named inventors' personal rights formally conveyed to the assignee two years after the 1995 filing. Not a distress transfer.

3. 1998-12-23 (recorded) / effective 1998-09-29 — Reel 009662/0497

• Conveyance: MERGER
• Assignor: AFFYMETRIX, INC., A CORP. OF CA
• Assignee: AFFYMETRIX INC., A CORP. OF DE
• Correspondent: not retrieved
• Context: internal corporate reincorporation — California Affymetrix merged into its Delaware parent/holding entity. Squarely a change of corporate form, not a change of ownership. Recorded-date match (1998-12-23) is exact, so this reel/frame is the best-supported confirmation in the chain.

4. 2012-06-27 (recorded) — Reel 028465/0541

• Conveyance: SECURITY AGREEMENT
• Assignor: AFFYMETRIX, INC.
• Assignee: GENERAL ELECTRIC CAPITAL CORPORATION, AS AGENT (Massachusetts)
• Correspondent: not retrieved
• Context: securitization / collateral pledge — GE Capital taken as agent for a lending syndicate with a security interest in the Affymetrix patent estate. The patent is collateral, not sold; Affymetrix remains owner. Recorded date matches Google Patents exactly for this patent.

5. 2015-11-13 (recorded) / effective 2015-10-28 — Reel 037109/0132

• Conveyance: RELEASE BY SECURED PARTY
• Assignor: GENERAL ELECTRIC CAPITAL CORPORATION, AS AGENT
• Assignee: AFFYMETRIX, INC.
• Correspondent: not retrieved
• Context: lien release — the security interest is discharged and Affymetrix holds the patent unencumbered. This is the last recorded event, and it lands roughly six months after the patent's 2015-04-28 anticipated expiration.

No records beyond 2015-11-13. In particular, there is no assignment to any IP-holding LLC, no license record, and no other post-issuance transfer. Affymetrix (later Thermo Fisher) is the terminal recorded owner.

Timeline diagram

timeline
    title Ownership of US 5744305
    1989 : Priority filing
    1990 : Parent case 492462 filed
    1995 : Application filed
    1996 : Affymax assigns to Affymetrix
         : Inventors assign their rights
    1998 : Patent issued
         : Merger into Delaware corporation
    2012 : GE Capital security agreement
    2015 : Security interest released
         : Patent term expired
    2016 : Affymetrix acquired by Thermo Fisher

(The 1996 and 2016 entries reflect respectively the Google Patents event date and my training-data recollection of the Thermo Fisher acquisition, flagged above as unverified in-session. The 1989/1990 entries are from the specification's own Cross Reference; they are priority chain events, not recorded assignments, and are included only to orient the diagram.)

NPE / troll-pattern signals

Verdict

Operating-company assertion.

Justification: This is a pure operating-company chain from start to finish. The patent issued 1998-04-28 to Affymetrix, Inc.; the only recorded transferees in its entire life are Affymetrix, Inc. of California (merger into Affymetrix Inc. of Delaware, effective 1998-09-29, recorded 1998-12-23, Reel 009662/0497) and General Electric Capital Corporation, as agent, which held a security interest recorded 2012-06-27 (Reel 028465/0541) that was released 2015-11-13 (Reel 037109/0132) — a pledge, never a sale, and released six months after the patent expired on 2015-04-28. The patent was asserted by its original assignee, in its own name, against actual competitors (Affymetrix v. Hyseq, N.D. Cal. C 98-03192, filed 1998-08-18 on the '305 patent; Affymetrix v. Synteni/Incyte, settled 2001-12-21 by mutual cross-license), and Affymetrix shipped GeneChip® glass oligonucleotide arrays that read directly onto claims 1 and 15. There is no NPE, no shell LLC, no asserter and no aggregator anywhere in the 26-year record.

Two limits on this verdict, stated plainly: (a) the reel/frame numbers for links 1 and 2 (Affymax→Affymetrix; inventors→Affymetrix) are unretrieved and unconfirmed — only the 1998, 2012 and 2015 reel/frames are date-corroborated, and those are corroborated via a sibling patent's INPADOC record, not the '305 number; and (b) the correspondent-of-record field was not obtainable, so signal 3 is genuinely unclear rather than negative. Neither gap can change the verdict, because both the assignor and assignee side of every event are named operating entities and the terminal event is a lender's lien release.

Verify at: https://assignment.uspto.gov/patent/index.html — search by patent number 5744305 (also reachable via https://assignmentcenter.uspto.gov/). Compare reel/frame against 008955/0530, 009662/0497, 028465/0541, and 037109/0132, and capture the correspondent-of-record field that I could not reach.

I'll search for the specific prior-art citations of US 5,744,305.

Let me pull the actual citation lists from the patent's front page.

Let me get the front-page reference lists from Justia and FreePatentsOnline.

Method note and provenance (read first)

Two housekeeping items before the substantive analysis, because they affect how much weight to put on what follows:

1. Date discrepancy — flagged, not silently resolved. My operating environment reports today as 2026‑09‑10; the task header says April 26, 2026. I have not auto-corrected either. This does not change the analysis (the patent expired 2015‑04‑28), but per the cross-reference rule I am flagging the contradiction rather than papering over it.

2. I could not read US 5,744,305's own "References Cited" front‑page block verbatim in this session. The Google Patents document supplied for this task omits it entirely — the scrape runs bibliographic data → family/legal events → classifications → description, and the description is truncated mid‑sentence ("This cycle of coupling…"), before the claims. Targeted searches returned only fragments of the reference list (mostly non‑patent literature snippets from the uspto.report mirror) and the reference lists of sibling patents sharing the same specification (US 6,566,495; US 7,087,732). Accordingly:

• U.S. patent references below marked [carried forward] come from the previously generated section of this analysis, which I treat as authoritative for cross‑reference per instructions. I did not independently re‑verify that list this session.
• References marked [sibling list] come from the front page of US 6,566,495 / US 7,087,732, which are the same specification (continuations of 08/466,632 / 09/063,933). Their citation lists are highly likely but not proven to be identical to 5,744,305's.
• Nothing below has been auto‑corrected; where I give a date I am uncertain of, I say so.

1. The threshold question: what can even be §102 art against this patent

This is where most analyses of 5,744,305 go wrong, because the patent's front page lists fifty-odd references and it is tempting to rank them all as "prior art." Under pre‑AIA §102 (correct law here — the application was filed June 6, 1995), the analysis is date-gated.

Consequences:

• §102(a) / §102(b): a reference only anticipates if it was publicly available (or patented) before June 7, 1989 — or before March 7, 1990 for claim matter not supported by the 1989 application. Because the specification does support the array claims across the chain (it recites regions down to 10⁻¹⁰ cm² and "more than about 10³, 10⁴, 10⁵, 10⁶, 10⁷, or 10⁸ different sequences" on a single substrate), the June 7, 1989 date governs the anticipation analysis for claims 1 and 15.
• Therefore every post‑1989 reference cited on the face of this patent is functionally incapable of §102(a)/(b) anticipation. That includes the applicant's own later work (Fodor et al., Science 251:767‑773 (1991)), Wilcox et al. (1990), Veldkamp (CLEO 90, May 21, 1990), Pirrung et al. (1995), and the whole family of Affymax/Affymetrix continuation patents.
• §102(e): a U.S. patent or published application is prior art as of its own earliest effective U.S. filing date — so a 1993-issued patent with a 1988–89 filing date can be §102(e) art. This is the only door through which most of the cited U.S. patents could enter.
• Whole-family self-citations are not prior art. US 5,143,854, US 5,405,783, US 5,445,934 and US 5,489,678 are the parents/ancestors of 08/466,632. For the shared disclosure, 5,744,305 is entitled to the same effective date, so these cannot be §102 art against it (and §103(c) common-ownership would be an alternative bar). Their citation on the face of the patent reflects the examiner's need to establish the continuity record, not a prior-art rejection.

Net effect: the only references in the cited list with genuine §102 anticipation potential are those public before 1989‑06‑07, plus any U.S. patent/application with a pre‑1989 effective filing date under §102(e). Everything else is §103 (obviousness) fodder at best.

2. U.S. patent references — per-reference analysis

2a. Same-family references (cited, but legally not prior art)

Analyst's note (and the key doctrinal point for this patent): US 5,143,854 does disclose an array of >400 different predefined‑region polymers on a solid support — i.e., it describes everything claim 1 recites except, arguably, the "planar non‑porous" support limitation. If it were prior art it would be the single most dangerous reference in the file. Its non-prior-art status rests entirely on the continuity chain, which is exactly why the examiner documented that chain so thoroughly on the face of the patent.

2b. Third-party U.S. references [carried forward]

2c. Pre‑1989 U.S. patents [sibling list] — the genuine §102 window

These appear on the front page of the sibling US 6,566,495 (same specification, same examiner art unit lineage). If they also appear on 5,744,305's face page — which is likely — they are the references that actually sit inside the §102(b) window (issued before June 7, 1989, so more than one year before even the 1990 CIP):

3. Non-patent literature cited (fragments verified via the uspto.report mirror of the patent)

Bottom line on the NPL: not one of the cited non-patent references anticipates any of claims 1–26, individually or in substance. Their role is to establish that the building blocks (o‑nitrobenzyl photolabile groups, solid-phase synthesis, fluorescence detection, SOP for activated esters) were known — i.e., a §103 scaffolding for the combination, which is the actual inventive contribution.

4. Where the real §102 exposure to this patent actually sat: §102(g)

Because the patent's own cited art is almost entirely §103-grade, the substantive §102 attacks on US 5,744,305 in the real world came from §102(g)(1) — priority of invention by another, not from §102(a)/(e) documentary art:

• The Affymetrix–Incyte/Synteni interference, in which Incyte/Synteni sought to invalidate US 5,744,305 and US 5,800,992, is the only §102-caliber contest identified in the prior section of this analysis. Reported issues were claim construction of "discrete known regions," "oligonucleotides" vs. "polynucleotides" (claims 1 vs. 15 here), and a §112 enablement attack on cDNA arrays. The case settled before trial; the interference appeals were still described as pending at the time of the retrospective I found.
• Caveat, stated plainly: I have not verified a docket number, reporter citation, or a primary‑source record of the interference count from a primary source in this session. I am reporting it as a §102(g) exposure of the right type and target claims, not as a citation‑grade fact.
• No 2026 Federal Circuit activity on 5,744,305 was found. As stated in the prior section, that is a search limitation, not an affirmative negative — though it is consistent with the 2015‑04‑28 expiration.

5. Ranking: most relevant prior art for US 5,744,305

Ordered by actual legal bite, not by citation order:

The single most important analytical conclusion: because the effective date for the array claims is June 7, 1989, and because essentially all of the chemically and optically relevant cited art either (a) postdates that date, (b) is the applicant's own family/continuation, or (c) discloses only one component of the claimed combination, no reference cited on the face of US 5,744,305 anticipates claims 1 or 15 under §102(a), (b), or (e) on the record I was able to assemble. The citation list functions as an obviousness and enablement record, not an anticipation record. The patent's genuine §102 vulnerability was always §102(g) — the Incyte/Synteni interference — not documentary art.

6. Explicit uncertainties and non-findings

I want to be unambiguous about the limits of this answer, consistent with the earlier section's uncertainty list:

1. The verbatim front-page "References Cited" list of US 5,744,305 was not retrievable in this session. The provided Google Patents text omits it; my searches surfaced fragments only. The U.S. patent entries marked [carried forward] rest on the prior section's list; those marked [sibling list] rest on US 6,566,495's front page (same specification, ~same art unit) and are inferred, not confirmed, to match 5,744,305's.
2. I did not verify the description, assignee, or dates of US 5,252,743, US 5,258,506, or US 5,079,600. I have deliberately not invented descriptions for them.
3. US 5,202,231's filing date was not verified, which is the single fact that determines whether it is §102(e) art. My "likely §103 rather than §102" conclusion is provisional on that.
4. Non-patent reference page numbers are partially OCR-garbled in the mirror (e.g., "J. 37:2281‑2285", "J. 55:1585‑1589"). I have reproduced the volume/page data as it appeared rather than inferring the journal names from memory.
5. No primary-source confirmation of the Incyte/Synteni docket, interference number, or outcome was obtained; item 4 in the prior section and §4 above stand on secondary reporting.
6. No 2026 CAFC activity confirmed or excluded — stated as a search limitation.
7. Per the strict rule: the patent number 5744305 was searched literally and no other number was substituted. Nothing in the reference identifiers above has been auto-corrected.

I'll ground the obviousness analysis in the actual prior-art references of record for this patent before applying § 103.

Obviousness Analysis — US 5,744,305 under 35 U.S.C. § 103

0. Grounding note (and one correction to the earlier section)

The earlier analysis flagged that claim language came from two secondary mirrors. My searches this session added a third independent mirror (patenthub.cn) whose claim-15 and claim-26 text matches the uspto.report text verbatim, including the distinctive "density exceeding 400 different polynucleotides/cm²" and "planar non-porous solid support" phrasing. That reduces, but does not eliminate, the claim-sourcing uncertainty from §5 of the prior section. Nothing in this session contradicts the earlier sections; the cited-art list in the prior section is confirmed except that I can now name the previously unnamed reference: US 5,079,600 is Schnur et al. (Justia secondary listing, Jan. 7, 1992) — I could not verify its subject matter from a primary source, so I do not rely on it below.

1. Legal framework

The application underlying US 5,744,305 was filed June 6, 1995, so pre-AIA 35 U.S.C. § 103(a) governs. The analysis follows Graham v. John Deere, 383 U.S. 1 (1966), as qualified by KSR Int'l v. Teleflex, 550 U.S. 398 (2007) — i.e., the rationales are not limited to express "teaching, suggestion, or motivation" in the references; a predictable combination, a simple substitution of one known element for another, or "use of a known technique to improve similar devices in the same way" suffices.

Critically, claims 1 and 15 are product claims to a physical article, not method claims. Obviousness therefore turns on whether the structure recited (support geometry, density, regionalization, sequence determinability, length) would have been obvious — not on whether the method of making it was novel. This matters because the prior art contains extensive, well-developed disclosure of the method.

2. The threshold issue: effective filing date — and why it is the whole ballgame

Under § 120 a claim receives the benefit of an earlier application only for subject matter supported under § 112 ¶ 1 in that application. The '305 chain runs: 07/362,901 (Jun. 7, 1989) → 07/492,462 (Mar. 7, 1990) → 07/624,120 (Dec. 6, 1990) → 08/390,272 (Feb. 16, 1995) → 08/466,632 (Jun. 6, 1995). There are at least three candidate critical dates for claims 1 and 15, and the outcome of the § 103 analysis flips depending on which applies:

The specification as issued asserts that "sequences of nucleic acids may be synthesized to establish DNA or RNA binding sequences," that "each of the sets may include protected members," and that attachment of a protecting group to the 5'-hydroxyl of a nucleoside prevents side reaction — indicating that as of the 1995 text, nucleotide synthesis was contemplated. But the recitations that distinguish claims 1/15 — "planar non-porous" support at ">400 different polynucleotides/cm²" — trace to disclosure in this 1995 specification ("in one specific embodiment, the regions are between about 10 × 10 µm and 500 × 500 µm"; "more than about 10³, 10⁴, 10⁵, 10⁶, 10⁷, or 10⁸ different sequences"). Whether that identical text appears in the Mar. 7, 1990 parent I cannot verify from the materials available to me. This is the single most important unverified fact in the analysis, and a real-world § 103 challenge would be fought here first. I analyze below primarily on the 1995 date (most conservative for the patentee in the sense that it maximizes available art) and note where the 1990 date would change the result.

Also note the § 103(c) overlay: several of the most on-point references — US 5,143,854 (Pirrung), US 5,252,743 (Barrett), US 5,445,934, US 5,424,186, US 5,489,678, and WO 90/15070 — are all Affymax/Affymetrix subject matter. If any of them qualifies as prior art only under § 102(e)/(f)/(g), it is disqualified for § 103 purposes under pre-AIA § 103(c) because it was commonly owned when the invention was made. That disqualification does not reach Fodor Science 1991, Khrapko 1989, Geysen 1984/87, Houghten 1985, Southern/Maskos, Dattagupta US 4,542,102, or Lowe US 4,562,157. An "obviousness combination" built on the commonly-owned patents alone is vulnerable; one built on the non-commonly-owned printed publications plus the non-commonly-owned patents is not.

3. Combination A (primary): Fodor Science 1991 + Khrapko 1989 and/or Southern/Maskos + Dattagupta US 4,542,102

Claim 1 recites: (a) planar non-porous solid support with at least a first surface; (b) >400 different oligonucleotides/cm²; (c) each in a different predefined region; (d) different determinable sequence; (e) ≥4 nucleotides.

Motivation to combine. Khrapko, Southern & Maskos (Chem. Abstr. 113:152979r (1990); Genomics 13:1008–1017 (1992)) and Drmanac (US 5,202,231) all disclose the use of arrays of many different, sequence-known oligonucleotides on a solid phase for sequencing/diagnostics — but produce them by conventional, one-at-a-time means. Their methods improve monotonically with the number of different sequences addressable per unit area, supplying a concrete, articulated design incentive to increase density (the KSR "design incentive/market force" rationale). Fodor/Pirrung supply the known technique — photolithography — which the art already used to miniaturize, and which the '305 specification itself says makes "a high degree of miniaturization… possible because the density of compounds is determined largely with regard to spatial addressability." Using a known miniaturization technique to increase the capacity of a known array type is the paradigm KSR case.

Substitution rationale. The step from peptide arrays to oligonucleotide arrays is a substitution of one known class of monomer for another in an otherwise identical solid-phase synthesis cycle — the classic predictable-variation rationale. The record supports this directly: (i) US 5,252,743 (Barrett) — commonly owned, but confirming the state of the art — lists "oligonucleotides" and "polysaccharides" alongside antibodies and hormones as the anti-ligands to be immobilized on a spatially irradiated surface, i.e., the art treated peptide- and nucleic-acid-bearing surfaces as a single design space; (ii) the '305 specification itself relies on the same photolabile protecting groups (NVOC, NPOC, MeNVOC, MeNPOC, PyROC — see the specification's §"protecting groups") for both amino-terminus and 5'-hydroxyl protection, disclosing that "attachment of a protecting group to the 5'-hydroxyl group of a nucleoside… prevents the 5'-hydroxyl of one nucleoside from reacting with the 3'-activated phosphate-triester of another"; and (iii) the photolabile-group chemistry was itself long-known and non-commonly-owned: Patchornik, JACS 92:6333 (1970), Amit, J. Org. Chem. 39:192 (1974), McCray, Ann. Rev. Biophys. Biophys. Chem. 18:239–270 (1989) — all cited on the face of the family.

4. Combination B (alternative primary): Drmanac US 5,202,231 + Pirrung US 5,143,854 / Fodor Science

Drmanac (US 5,202,231, Apr. 13, 1993) is the non-commonly-owned sequencing-by-hybridization patent and is therefore § 103(c)-clean. It teaches the array-of-immobilized-oligonucleotides concept and (as reflected in the Drmanac-derived literature) the use of short probes of defined sequence. Combined with Pirrung/Fodor's mask-directed photodeprotection and coupling cycle, every element of claims 1 and 15 is accounted for. The motivation is the same: Drmanac's method demands a large, known-position set of oligos; light-directed synthesis was known to deliver exactly that, with the additional benefit that "position defines sequence," which is the only economically viable way to reach the >400/cm² and 1,000 / 10,000-sequence thresholds recited in dependent claims 17–18 and 23–25.

5. Combination C: Geysen/Houghten + Fodor/Pirrung ("miniaturize the known parallel format")

• Geysen et al., PNAS 81:3998–4002 (1984) and J. Immunol. Methods 102:259–274 (1987): parallel synthesis of many different peptides at physically discrete, known locations ("to a resolution of a single amino acid"), in a format arrayed against microtiter wells.
• Houghten, PNAS 82:5131–5135 (1985) ("tea-bag") and Houghten et al., Nature 354:84–86 (1991): large-number combinatorial synthesis on discrete solid-phase supports.
• Fodor/Pirrung: photocleavable protecting groups + mask-directed illumination on a planar support.

Motivation: the known formats (pins, tea bags) are intrinsically limited in feature density by mechanical handling; photolithography is a known technique for manufacturing dense, precisely registered patterns (the patent's own classification includes G03F 7/00 photomechanical patterning and G03F 7/0045). Applying it to a known parallel-synthesis format to improve density "in the same way" is a textbook KSR rationale. The "planar non-porous" limitation is what distinguishes this route from Geysen/Houghten's porous resins, and a flat glass or silica surface was the ordinary choice where optical readout of binding is the goal (Barrett; Dattagupta; the patent's fluorescent-scan Example).

6. The dependent claims

Claims 14 and 26 — product-by-process claims reciting "provide a support with protected compounds; deprotect a first portion but not a second; couple; deprotect a third portion comprising a fraction of the first portion; couple; optionally repeat" — recite precisely the binary/recursive masking described in Fodor/Pirrung and in this specification's own binary synthesis section ("a given cycle illuminates only about half of the region which was illuminated in a previous cycle, while protecting the remaining half… masking is recursive"). For § 103 these process steps add nothing: the binary masking strategy, the gray-code variant, and the modified gray-code variant are all disclosed as design choices within the same known masking art.

Claim 15 (the "polynucleotide" mirror of claim 1) is obvious for the identical reasons. The earlier section's note that Incyte argued "polynucleotide" should be limited to naturally occurring, longer-than-oligo DNA — and lost on plain meaning — is consistent with this: a claim whose only distinction is word choice over claim 1 is a fortiori obvious over the same art.

7. Anticipated counterarguments and how they fare

1. Teaching away. Barrett's discussion notes that prior photochemical nucleic-acid attachment methods "halve] low quantum yields in protic solvents, lacks spatial directability." But that criticism is aimed at a specific psoralen-mediated coupling of pre-formed nucleic acids (Dattagupta US 4,542,102; Lowe US 4,562,157), not at photolabile protecting groups used in in-situ chain assembly. The distinction is important: a teaching away must be directed to the claimed subject matter. Additionally, an internal Affymax criticism of a competitor approach does not implicate the § 103(c)-clean publications (Khrapko, Southern, Geysen, Houghten).
2. Unexpected results. The patent's strongest empirical assertions — MeNVOC/MeNPOC photolyze faster than NVOC/NPOC, and NVOC's nitrosobenzaldehyde by-product forms only a reversible Schiff base rather than poisoning the chain — are asserted in the specification and tabulated. However, these results bear on the protecting-group subject matter, not on claims 1 and 15. The array claims do not recite NVOC, MeNVOC, or any protecting group. Any rebuttal evidence tied to those groups is not commensurate in scope with the claims (cf. In re Hiniker).
3. "At the time of invention" (1989) argument. If the patentee successfully establishes a June 7, 1989 or Mar. 7, 1990 effective date for claims 1 and 15, the Fodor Science paper (Feb. 15, 1991) falls out as art, and the case for § 103 weakens appreciably — though Khrapko 1989, Geysen 1984/87, Houghten 1985, Dattagupta 1985, and Lowe 1985 survive. This is the decisive factual battleground.
4. § 103(c) common ownership. As noted, an obviousness case resting on US 5,143,854 / 5,252,743 / 5,445,934 / WO 90/15070 would be attacked as barred by § 103(c). The non-commonly-owned literature avoids that problem.
5. Enablement of cDNA arrays. The Incyte § 112 enablement attack (per the microarray IP retrospective in search results) is a § 112 defense, not a § 103 defense, and does not bear on the analysis here except as an impeachment of the patentee's asserted breadth.

8. Bottom line

On the assumption that claims 1 and 15 are entitled to no earlier than the 1995 filing date, both independent claims would very likely have been held obvious under § 103(a) over:

• Fodor et al., Science 251:767–773 (1991) (or Pirrung US 5,143,854 for the method disclosure, subject to § 103(c)), as the primary reference teaching mask-directed, light-directed parallel synthesis on a planar solid support with known, positionally determined sequences at lithographically scalable density; in view of
• Khrapko et al., FEBS Lett. 256:118–122 (1989) and/or Southern & Maskos (1990/1992) and/or Drmanac US 5,202,231, teaching arrays of many different, sequence-known oligonucleotides on a solid support for hybridization analysis and supplying the explicit motivation to increase the number of different sequences per unit area; optionally further in view of
• Dattagupta US 4,542,102 and/or Barrett US 5,252,743 for attaching nucleic acids to a planar, non-porous solid support through a linker, and Geysen 1984/87 / Houghten 1985 for the known parallel-synthesis-at-defined-locations format to be improved by the known photolithographic technique.

The dependent claims add only conventional design choices (length, density, separation, glass, linker, purity thresholds) that the art and the specification itself recite as within the ordinary range, and are obvious for the same reasons.

The claims survive only if the patentee wins the priority fight — i.e., establishes that the 1989/1990 applications provide § 112 ¶ 1 support for a "planar non-porous solid support" bearing ">400 different oligonucleotides/cm²," which in turn would knock out the 1991 Fodor Science publication. Given the claims' express recitation of nucleotide (not peptide) arrays and micrometer-scale density, and given that the earliest application (07/362,901) is characterized in the family as the VLSIPS polypeptide work, I regard that showing as unlikely but not foreclosed.

No 2026 CAFC docket activity on 5,744,305 was found in this session — consistent with the 2015-04-28 expiration, and stated as "no evidence found," not as a confirmed negative.